It seems you may have completely misunderstood my post (and you missed the <sarcasm> tags I kept putting in there at least once); perhaps I was too vague in how I answered. I'll give it my best shot and try to clarify my meaning.
You say that people would set up the servers for us with open standards. They do that now, and many with sophisticated turnkey solutions that completely remove the learning curve from the equation (e.g. squarespace.com) What are you proposing we add that can improve on the open standards we already have?
These services are defeated by Facebook because they're decentralized; people want go to Facebook on the internet in the same way that they go to their favorite bar in real life. They crave a group social aspect that the open-standard model proposed here would find difficult to replace. The reason Facebook is successful is not because everybody's already on it, but because it offers a sense of commonality that people like to snuggle up in like a security blanket. "Separate but equal" doesn't jibe when what people really want is to just be together.
Email is a necessary protocol. It's a fundamental messaging system that it was more pragmatic to open up to an open standard than it was to maintain as a proprietary protocol; companies really had something to lose if they tried to lock their users away from the rest of the world. Comparing the functionality of email to Facebook is, to again use the mundane world example, like comparing sending a letter to somebody versus going to meet them at everyone's favorite hangout. Once again, not a lot in common. While everyone can agree that a basic form of correspondence is a necessity and doesn't necessarily call for a lot of bells and whistles to get the job done, people choose their social scene based on a different set of criteria, where often the more bells and whistles can be offered the better. One of the major drawbacks of open standards is that adding new features can be a long time in coming, and they're often beaten to the punch years ahead by a closed-source solution while the standard is waiting to be finalized (Flash vs. HTML 5, for example).
The problem with fragmentation is that there is little way to communicate between them. If this problem was solved, fragmentation wouldn't be a problem. Yet, open standards meant that email got far more fragmented, but the point is that [this] isn't a problem.
Okay, this part of your response is a bit confusing. At first glance, and second, it almost seems like you're agreeing with me. You're saying that fragmentation wouldn't be a problem if it was solved, which I think we can all agree with. However, you don't offer any solution to solving it. I agree; there are a lot of problems in this world that wouldn't be so bad if only they could be solved. Saying it doesn't make it go away, though.
Also, you're saying that opening up email led to fragmentation, which is my point exactly, so how is this not a problem? (Just an FYI, email hasn't suffered from any major fragmentation issues since the late nineties, so I think it's fair to say that we can let that one go).
"The web doesn't solve the problems we are discussing here.
Yes, I know. That was my point. I'm sorry you missed the subtext there. It doesn't solve the problem we have, and adding yet another protocol to solve the problem is only going to add to the confusion, not alleviate the problem.
What has desktop computing got to do with this?
Okay, that was the sarcasm part; please go back and read it again with the proper inflection. My point was that Linux did nothing to displace the proprietary software platforms out there that it claimed superiority to (and we're talking about platforms here, not software, so the comparison is still valid). I seriously doubt an open standard for social networking is going to be any more successful.
Perhaps this would have been a better analogy if I'd used the obligatory car reference; sorry for having gone outside the box on this one. Better luck next time, huh?